xen-devel.lists.xenproject.org archive mirror
 help / color / mirror / Atom feed
* [PATCH for-4.7] Revert "libxl: No emulated disk driver for xvdX disk"
@ 2016-06-08 15:31 Wei Liu
  2016-06-08 15:32 ` Ian Jackson
  0 siblings, 1 reply; 10+ messages in thread
From: Wei Liu @ 2016-06-08 15:31 UTC (permalink / raw)
  To: Xen-devel; +Cc: Anthony PERARD, George Dunlap, Ian Jackson, Wei Liu

This reverts c0c099d157cc5bc942afef766cf141628a6380a1.

That commit causes regression on the semantics of our diskspec.
See [0] for more information.

[0] http://lists.xen.org/archives/html/xen-devel/2016-05/msg02876.html

Signed-off-by: Wei Liu <wei.liu2@citrix.com>
---
Please review carefully about this patch. There are one commit that is
of interest: ef6cb76026628e26e3d1ae53c50ccde1c3c78b1b

I believe the reverting the snippet in question won't cause security
issue as described in XSA-142, because the code to create IDE disk still
checks if the disk is read-only.
---
 tools/libxl/libxl_dm.c | 6 ------
 1 file changed, 6 deletions(-)

diff --git a/tools/libxl/libxl_dm.c b/tools/libxl/libxl_dm.c
index 155a653..6ff05c3 100644
--- a/tools/libxl/libxl_dm.c
+++ b/tools/libxl/libxl_dm.c
@@ -1422,12 +1422,6 @@ static int libxl__build_device_model_args_new(libxl__gc *gc,
                                                         format,
                                                         &disks[i],
                                                         colo_mode);
-                } else if (strncmp(disks[i].vdev, "xvd", 3) == 0) {
-                    /*
-                     * Do not add any emulated disk when PV disk are
-                     * explicitly asked for.
-                     */
-                    continue;
                 } else if (disk < 6 && b_info->u.hvm.hdtype == LIBXL_HDTYPE_AHCI) {
                     if (!disks[i].readwrite) {
                         LOG(ERROR, "qemu-xen doesn't support read-only AHCI disk drivers");
-- 
2.1.4


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: [PATCH for-4.7] Revert "libxl: No emulated disk driver for xvdX disk"
  2016-06-08 15:31 [PATCH for-4.7] Revert "libxl: No emulated disk driver for xvdX disk" Wei Liu
@ 2016-06-08 15:32 ` Ian Jackson
  2016-06-08 15:52   ` Wei Liu
  0 siblings, 1 reply; 10+ messages in thread
From: Ian Jackson @ 2016-06-08 15:32 UTC (permalink / raw)
  To: Wei Liu; +Cc: Anthony PERARD, Xen-devel, George Dunlap

Wei Liu writes ("[PATCH for-4.7] Revert "libxl: No emulated disk driver for xvdX disk""):
> This reverts c0c099d157cc5bc942afef766cf141628a6380a1.
> 
> That commit causes regression on the semantics of our diskspec.
> See [0] for more information.
> 
> [0] http://lists.xen.org/archives/html/xen-devel/2016-05/msg02876.html
> 
> Signed-off-by: Wei Liu <wei.liu2@citrix.com>
> ---
> Please review carefully about this patch. There are one commit that is
> of interest: ef6cb76026628e26e3d1ae53c50ccde1c3c78b1b
> 
> I believe the reverting the snippet in question won't cause security
> issue as described in XSA-142, because the code to create IDE disk still
> checks if the disk is read-only.

Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>

I think this is the right answer for 4.7.  In 4.8 we should do
something like one of the proposals being discussed in this thread.

Thanks,
Ian.

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: [PATCH for-4.7] Revert "libxl: No emulated disk driver for xvdX disk"
  2016-06-08 15:32 ` Ian Jackson
@ 2016-06-08 15:52   ` Wei Liu
  2016-06-08 16:00     ` George Dunlap
  0 siblings, 1 reply; 10+ messages in thread
From: Wei Liu @ 2016-06-08 15:52 UTC (permalink / raw)
  To: Ian Jackson; +Cc: Anthony PERARD, Xen-devel, Wei Liu, George Dunlap

On Wed, Jun 08, 2016 at 04:32:13PM +0100, Ian Jackson wrote:
> Wei Liu writes ("[PATCH for-4.7] Revert "libxl: No emulated disk driver for xvdX disk""):
> > This reverts c0c099d157cc5bc942afef766cf141628a6380a1.
> > 
> > That commit causes regression on the semantics of our diskspec.
> > See [0] for more information.
> > 
> > [0] http://lists.xen.org/archives/html/xen-devel/2016-05/msg02876.html
> > 
> > Signed-off-by: Wei Liu <wei.liu2@citrix.com>
> > ---
> > Please review carefully about this patch. There are one commit that is
> > of interest: ef6cb76026628e26e3d1ae53c50ccde1c3c78b1b
> > 
> > I believe the reverting the snippet in question won't cause security
> > issue as described in XSA-142, because the code to create IDE disk still
> > checks if the disk is read-only.
> 
> Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
> 
> I think this is the right answer for 4.7.  In 4.8 we should do
> something like one of the proposals being discussed in this thread.
> 

Thanks.  And Anthony said on IRC this patch looked good to him.

I've queued this for committing to both unstable and 4.7.

Wei.

> Thanks,
> Ian.

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: [PATCH for-4.7] Revert "libxl: No emulated disk driver for xvdX disk"
  2016-06-08 15:52   ` Wei Liu
@ 2016-06-08 16:00     ` George Dunlap
  2016-06-08 16:17       ` Olaf Hering
  0 siblings, 1 reply; 10+ messages in thread
From: George Dunlap @ 2016-06-08 16:00 UTC (permalink / raw)
  To: Wei Liu, Ian Jackson
  Cc: Anthony PERARD, Xen-devel, Olaf Hering, George Dunlap

On 08/06/16 16:52, Wei Liu wrote:
> On Wed, Jun 08, 2016 at 04:32:13PM +0100, Ian Jackson wrote:
>> Wei Liu writes ("[PATCH for-4.7] Revert "libxl: No emulated disk driver for xvdX disk""):
>>> This reverts c0c099d157cc5bc942afef766cf141628a6380a1.
>>>
>>> That commit causes regression on the semantics of our diskspec.
>>> See [0] for more information.
>>>
>>> [0] http://lists.xen.org/archives/html/xen-devel/2016-05/msg02876.html
>>>
>>> Signed-off-by: Wei Liu <wei.liu2@citrix.com>
>>> ---
>>> Please review carefully about this patch. There are one commit that is
>>> of interest: ef6cb76026628e26e3d1ae53c50ccde1c3c78b1b
>>>
>>> I believe the reverting the snippet in question won't cause security
>>> issue as described in XSA-142, because the code to create IDE disk still
>>> checks if the disk is read-only.
>>
>> Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
>>
>> I think this is the right answer for 4.7.  In 4.8 we should do
>> something like one of the proposals being discussed in this thread.
>>
> 
> Thanks.  And Anthony said on IRC this patch looked good to him.
> 
> I've queued this for committing to both unstable and 4.7.

CC'ing Olaf and Konrad for their information. :-)

 -George


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: [PATCH for-4.7] Revert "libxl: No emulated disk driver for xvdX disk"
  2016-06-08 16:00     ` George Dunlap
@ 2016-06-08 16:17       ` Olaf Hering
  2016-06-08 17:55         ` Konrad Rzeszutek Wilk
  0 siblings, 1 reply; 10+ messages in thread
From: Olaf Hering @ 2016-06-08 16:17 UTC (permalink / raw)
  To: George Dunlap
  Cc: Wei Liu, George Dunlap, Ian Jackson, Anthony PERARD, Xen-devel

On Wed, Jun 08, George Dunlap wrote:

> CC'ing Olaf and Konrad for their information. :-)

I'm fine with the revert.

Olaf

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: [PATCH for-4.7] Revert "libxl: No emulated disk driver for xvdX disk"
  2016-06-08 16:17       ` Olaf Hering
@ 2016-06-08 17:55         ` Konrad Rzeszutek Wilk
  2016-06-09 13:23           ` Olaf Hering
  0 siblings, 1 reply; 10+ messages in thread
From: Konrad Rzeszutek Wilk @ 2016-06-08 17:55 UTC (permalink / raw)
  To: Olaf Hering
  Cc: Wei Liu, George Dunlap, Ian Jackson, George Dunlap,
	Anthony PERARD, Xen-devel

On Wed, Jun 08, 2016 at 06:17:55PM +0200, Olaf Hering wrote:
> On Wed, Jun 08, George Dunlap wrote:
> 
> > CC'ing Olaf and Konrad for their information. :-)
> 
> I'm fine with the revert.

Me too!

Thanks!
> 
> Olaf

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: [PATCH for-4.7] Revert "libxl: No emulated disk driver for xvdX disk"
  2016-06-08 17:55         ` Konrad Rzeszutek Wilk
@ 2016-06-09 13:23           ` Olaf Hering
  2016-06-13  8:41             ` George Dunlap
  0 siblings, 1 reply; 10+ messages in thread
From: Olaf Hering @ 2016-06-09 13:23 UTC (permalink / raw)
  To: Konrad Rzeszutek Wilk
  Cc: Wei Liu, George Dunlap, Ian Jackson, George Dunlap,
	Anthony PERARD, Xen-devel

On Wed, Jun 08, Konrad Rzeszutek Wilk wrote:

> On Wed, Jun 08, 2016 at 06:17:55PM +0200, Olaf Hering wrote:
> > On Wed, Jun 08, George Dunlap wrote:
> > 
> > > CC'ing Olaf and Konrad for their information. :-)
> > 
> > I'm fine with the revert.

Another case is pvgrub2, which also uses the 'dev' property to name the
disk internally "xen/$vdev". Not sure if any generated grub.cfg actually
contains tne name. But now that its reverted everything should continue
to work as it did in xen-4.6.

Olaf

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: [PATCH for-4.7] Revert "libxl: No emulated disk driver for xvdX disk"
  2016-06-09 13:23           ` Olaf Hering
@ 2016-06-13  8:41             ` George Dunlap
  2016-06-13 10:21               ` Olaf Hering
  0 siblings, 1 reply; 10+ messages in thread
From: George Dunlap @ 2016-06-13  8:41 UTC (permalink / raw)
  To: Olaf Hering; +Cc: Anthony PERARD, Ian Jackson, Wei Liu, Xen-devel

On Thu, Jun 9, 2016 at 2:23 PM, Olaf Hering <olaf@aepfle.de> wrote:
> On Wed, Jun 08, Konrad Rzeszutek Wilk wrote:
>
>> On Wed, Jun 08, 2016 at 06:17:55PM +0200, Olaf Hering wrote:
>> > On Wed, Jun 08, George Dunlap wrote:
>> >
>> > > CC'ing Olaf and Konrad for their information. :-)
>> >
>> > I'm fine with the revert.
>
> Another case is pvgrub2, which also uses the 'dev' property to name the
> disk internally "xen/$vdev". Not sure if any generated grub.cfg actually
> contains tne name. But now that its reverted everything should continue
> to work as it did in xen-4.6.

Ah, so pvgrub2 follows the xenolinux code, rather than the pvops code?
 That should definitely be changed ASAP.

 -George

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: [PATCH for-4.7] Revert "libxl: No emulated disk driver for xvdX disk"
  2016-06-13  8:41             ` George Dunlap
@ 2016-06-13 10:21               ` Olaf Hering
  2016-06-13 10:23                 ` George Dunlap
  0 siblings, 1 reply; 10+ messages in thread
From: Olaf Hering @ 2016-06-13 10:21 UTC (permalink / raw)
  To: George Dunlap; +Cc: Anthony PERARD, Ian Jackson, Wei Liu, Xen-devel

On Mon, Jun 13, George Dunlap wrote:

> Ah, so pvgrub2 follows the xenolinux code, rather than the pvops code?
>  That should definitely be changed ASAP.

What needs to be changed? It has to construct some internal identifier,
and appearently "xen/$vdev" was the most obvious choice.

Olaf

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: [PATCH for-4.7] Revert "libxl: No emulated disk driver for xvdX disk"
  2016-06-13 10:21               ` Olaf Hering
@ 2016-06-13 10:23                 ` George Dunlap
  0 siblings, 0 replies; 10+ messages in thread
From: George Dunlap @ 2016-06-13 10:23 UTC (permalink / raw)
  To: Olaf Hering, George Dunlap
  Cc: Anthony PERARD, Ian Jackson, Wei Liu, Xen-devel

On 13/06/16 11:21, Olaf Hering wrote:
> On Mon, Jun 13, George Dunlap wrote:
> 
>> Ah, so pvgrub2 follows the xenolinux code, rather than the pvops code?
>>  That should definitely be changed ASAP.
> 
> What needs to be changed? It has to construct some internal identifier,
> and appearently "xen/$vdev" was the most obvious choice.

I took you to mean that the on-disk grub.cfg would contain references to
"hda" rather than "xvda", which would be problematic.  If there's no
user-visible effect of this choice then of course it doesn't matter in
the slightest (and bringing it up was a red herring).

 -George

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel

^ permalink raw reply	[flat|nested] 10+ messages in thread

end of thread, other threads:[~2016-06-13 10:25 UTC | newest]

Thread overview: 10+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2016-06-08 15:31 [PATCH for-4.7] Revert "libxl: No emulated disk driver for xvdX disk" Wei Liu
2016-06-08 15:32 ` Ian Jackson
2016-06-08 15:52   ` Wei Liu
2016-06-08 16:00     ` George Dunlap
2016-06-08 16:17       ` Olaf Hering
2016-06-08 17:55         ` Konrad Rzeszutek Wilk
2016-06-09 13:23           ` Olaf Hering
2016-06-13  8:41             ` George Dunlap
2016-06-13 10:21               ` Olaf Hering
2016-06-13 10:23                 ` George Dunlap

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).